CN108665333A - Method of Commodity Recommendation, device, electronic equipment and storage medium - Google Patents

Method of Commodity Recommendation, device, electronic equipment and storage medium Download PDF

Info

Publication number
CN108665333A
CN108665333A CN201710206383.1A CN201710206383A CN108665333A CN 108665333 A CN108665333 A CN 108665333A CN 201710206383 A CN201710206383 A CN 201710206383A CN 108665333 A CN108665333 A CN 108665333A
Authority
CN
China
Prior art keywords
user
commodity
behavior
data
shopping
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201710206383.1A
Other languages
Chinese (zh)
Other versions
CN108665333B (en
Inventor
马兴国
高阳
董小平
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Jingdong Century Trading Co Ltd
Beijing Jingdong Shangke Information Technology Co Ltd
Original Assignee
Beijing Jingdong Century Trading Co Ltd
Beijing Jingdong Shangke Information Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beijing Jingdong Century Trading Co Ltd, Beijing Jingdong Shangke Information Technology Co Ltd filed Critical Beijing Jingdong Century Trading Co Ltd
Priority to CN201710206383.1A priority Critical patent/CN108665333B/en
Publication of CN108665333A publication Critical patent/CN108665333A/en
Application granted granted Critical
Publication of CN108665333B publication Critical patent/CN108665333B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions
    • G06Q30/0601—Electronic shopping [e-shopping]
    • G06Q30/0631—Item recommendations
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0201—Market modelling; Market analysis; Collecting market data

Landscapes

  • Business, Economics & Management (AREA)
  • Finance (AREA)
  • Accounting & Taxation (AREA)
  • Engineering & Computer Science (AREA)
  • Development Economics (AREA)
  • Strategic Management (AREA)
  • General Business, Economics & Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Game Theory and Decision Science (AREA)
  • Data Mining & Analysis (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

A kind of Method of Commodity Recommendation of present invention offer, system, electronic equipment and storage medium, commodity association degree can be calculated based on the characteristic of user behavior and user does shopping and is intended to commodity, to be user's Recommendations in the complexity of reduction commodity association degree calculating and in view of user's browsing compares the behavioral characteristic of commodity.This method comprises the following steps：Behavioral data is obtained to the behavior of commodity to be selected from user；Based on the behavioral data in the offline shopping period, the degree of association of commodity is calculated；In the commodity set that the degree of association is more than predetermined threshold, based on the behavioral data in the online shopping period, commodity, the commodity as user to be recommended to are bought to calculate user view.

Description

Commodity recommendation method and device, electronic equipment and storage medium
Technical Field
The invention relates to a commodity recommendation method, a commodity recommendation system, electronic equipment and a storage medium based on user behavior sequence mining.
Background
In recent years, electronic commerce has been promoted, and more users purchase desired goods by selecting online shopping. With the continuous expansion of the electronic commerce scale, the variety and the number of commodities rapidly increase, and a user needs to spend a lot of time to find out a needed commodity from a large quantity of commodities. In order to improve the shopping experience of the user, the shopping website provides personalized decision support and commodity information service for the user through a recommendation system.
The recommendation system includes three modules: the system comprises a user interest module, a recommendation object module and a recommendation algorithm module. The recommendation system matches the user interest information with the information in the recommended objects, and meanwhile, corresponding recommendation algorithms are used for calculation and screening, so that the recommended objects which are possibly interested by the user are found and then recommended to the user. The core of the user interest module is to calculate the purchase intention goods of the user. The core of the recommendation object module is to calculate the associated commodities of each commodity. The core of the recommendation algorithm module is to input the purchase intention commodities, recall the associated commodities from the recommendation object module and then reasonably sort the recall results.
In the prior art, a common way to calculate the shopping intention commodities is to take all commodities browsed by a user in the recent period of time as a shopping intention commodity set. The general way of calculating the association degree of the commodity is to decompose the commodity title information into feature vectors in a word segmentation mode, and the association degree of the commodity is expressed by the distance of the feature vectors. However, the shopping behavior of the user is characterized by browsing the comparison commodities and making a decision according to the comparison information so as to select the required commodity. The recently browsed commodity set is directly used as the shopping intention, the behavior characteristic of the user is not considered, and redundant comparison commodity data exist in the shopping intention commodities. In addition, the process of extracting the feature vector from the commodity title relates to Chinese word segmentation, and the Chinese word segmentation technology has higher development cost. Meanwhile, the calculation of the feature vector distance may face a high-dimensional feature space, and the complexity is high during calculation.
Disclosure of Invention
In view of this, embodiments of the present invention provide a commodity recommendation method, system, electronic device and storage medium based on user behavior sequence mining, which can provide personalized commodity recommendation for a user by means of a commodity association degree that reduces computational complexity and improves reliability, and a commodity with a shopping intention calculated in consideration of a characteristic that the user makes a purchase after browsing and comparing commodities.
To achieve the above object, according to an aspect of the embodiments of the present invention, a commodity recommendation method based on user behavior sequence mining is provided.
The commodity recommendation method according to the present invention includes the steps of: behavior data are obtained from behaviors of a user on a commodity to be selected and collected into a user behavior log in real time, wherein the behavior data comprise behavior data in an offline shopping period and behavior data in an online shopping period; calculating the association degree of the commodity based on the behavior data in the offline shopping period; and calculating the intention of the user to purchase the commodity as the commodity to be recommended to the user based on the behavior data in the online shopping period in the commodity set with the relevance degree larger than the preset threshold value.
Optionally, the behavior data includes: search, browse, collect, join shopping cart, and purchase.
The method for calculating the commodity association degree based on the offline shopping cycle and by adopting the idea of frequent pattern mining comprises the following steps: according to the behavior type of the user, extracting browsing data of the user and purchasing data of the user from behavior data in an offline shopping period; counting any commodity I in the browsing data of the useriN; performing correlation calculation on browsing data of the user and purchasing data of the user to determine similarity or matching degree of similar commodities, wherein the similarity of similar commodities refers to browsing of the commodity I by the user in one visit from logging in to exiting from a shopping websiteiBut purchase the same purpose goods IjThe commodity collocation degree refers to that the user browses the commodity I from one visit of logging in to exiting the shopping websiteiBut purchase goods I of a different purposejThe case (1); counting any commodity I according to the result of the previous stepiWith the purchased goods IjThe number of co-occurrences K; according to the formula R (I)i,Ij) Calculating the Commodity I as K/NiAnd IjI.e. browsing the item IiBut purchase the goods IjThe probability of (c).
Optionally, the offline shopping period is one month.
Calculating the shopping intent items based on the online shopping period, for example, includes the steps of: obtaining a user behavior sequence in an online shopping period from the user behavior log; constructing the following preference function Pre (I) of the final purchased commodity I under the condition of considering the frequency of the final purchased commodity I appearing in the user behavior sequence, the duration of the final purchased commodity I behavior performed by the user and the type of the user behavior, and assuming that { S } isiThe length of the sequence is N, and the last purchased commodity I is in the sequence SiThe set of positions in is { P }i}，1≤PiN is less than or equal to N, the weight factor of the behavior type is { Wi}，Where Σ Wi＝1，
Wherein wpRepresenting the behavior weight corresponding to the position p;
and calculating the preference value of the last purchased commodity I in the user behavior sequence in the online shopping period, comparing the preference value with a preset preference threshold value, and selecting the commodity with the preference value larger than the preset preference threshold value as the shopping intention commodity of the user.
To achieve the above object, according to another aspect of an embodiment of the present invention, there is provided an article recommendation device.
The commodity recommending device of the embodiment of the invention comprises: the system comprises an acquisition behavior data module, a recommendation object module and a user interest module, wherein the acquisition behavior data module is used for acquiring behavior data from the behavior of a user on a commodity to be selected and collecting the behavior data into a user behavior log in real time; the recommendation object module calculates commodity association degree by adopting a thought of frequent pattern mining based on an offline shopping period; the user interest module calculates, among a set of commodities whose degree of association is greater than a predetermined threshold, a commodity that the user intends to purchase as a commodity to be recommended to the user based on behavior data within an online shopping period.
Optionally, the behavioral data is data regarding at least one of searching, browsing, collecting, joining a shopping cart, and purchasing.
The method for calculating the association degree of the commodities in the commodity recommending device comprises the following steps: according to the behavior type of the user, extracting browsing data of the user and the purchase number of the user from the behavior data in the offline shopping periodAccordingly; counting any commodity I in the browsing data of the useriN; performing correlation calculation on the browsing data of the user and the purchasing data of the user to determine the similarity of the commodities of the same category or the matching degree of the commodities, wherein the similarity of the commodities of the same category refers to the commodity I browsed by the user from logging in to exiting the shopping website in one visitiBut purchase the same purpose goods IjThe commodity collocation degree refers to that a user browses the commodity I from one visit of logging in to quitting the shopping websiteiBut purchase goods I of a different purposejThe case (1); counting any commodity I according to the result of the correlation calculationiWith the purchased goods IjThe number of co-occurrences K; and according to the formula R (I)i,Ij) Calculating the Commodity I as K/NiAnd IjAs the item I being browsediThereafter purchasing goods IjThe possibility of (a).
Optionally, the offline shopping period is one month.
The method for calculating the intention of the user to purchase the commodity in the commodity recommending device comprises the following steps: obtaining a user behavior sequence in an online shopping period from the user behavior log; the following preference function Pre (I) for the last purchased item I is constructed, assuming { S }iThe length of the sequence is N, and the last purchased commodity I is in the sequence SiThe set of positions in is { P }i}，1≤PiN is less than or equal to N, the weight factor of the behavior type is { WiIn which Σ Wi＝1，
Wherein wpRepresenting the behavior weight corresponding to the position p;
and finally, comparing the calculated preference value with a preference threshold value and taking the commodity with the preference value larger than the preference threshold value as a commodity which is intended to be purchased by the user.
To achieve the above object, according to still another aspect of embodiments of the present invention, there is provided an electronic apparatus.
An electronic device of an embodiment of the present invention includes: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform a method for merchandise recommendation mined based on a sequence of user behaviors according to an embodiment of the present invention.
To achieve the above object, according to still another aspect of an embodiment of the present invention, there is provided a computer-readable storage medium.
A computer-readable storage medium of an embodiment of the present invention stores computer instructions for causing a computer to execute a commodity recommendation method mined based on a user behavior sequence of an embodiment of the present invention.
One embodiment of the above invention has the following advantages or benefits: the commodity association degree is mined according to the behavior mode which is possible to buy after the user browses and compares commodities in consideration of the behavior sequence of user shopping, so that the complexity of calculating the association degree is reduced, and the reliability of the association degree is improved. And (4) providing a calculation formula of the shopping intention commodity by mining the real-time behavior sequence of the user. The calculation formula based on the statistical result can better reflect the true shopping intention of the user. On the basis, commodity recommendation is carried out for the user in a personalized mode.
The above technical solutions will be described below with reference to specific embodiments.
Drawings
The drawings are included to provide a better understanding of the invention and are not to be construed as unduly limiting the invention. Wherein:
fig. 1 is a schematic flowchart of a commodity recommendation method mined based on a user behavior sequence according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of the main steps of calculating the association degree of the goods based on the offline shopping period and adopting the idea of frequent pattern mining according to the embodiment of the present invention;
FIG. 3 is a schematic diagram of the main steps of a method for calculating an intent-to-purchase item based on an online shopping period, according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of a merchandise recommendation device having an obtain behavior data module, a recommendation object module, and a user interest module according to an embodiment of the present invention;
fig. 5 is a schematic diagram of a hardware structure of an electronic device according to an embodiment of the present invention.
Detailed Description
Exemplary embodiments of the present invention are described below with reference to the accompanying drawings, in which various details of embodiments of the invention are included to assist understanding, and which are to be considered as merely exemplary. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the invention. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.
The method aims to solve the technical problems in the prior art that a Chinese word segmentation technology with high cost is designed by extracting a feature vector from a commodity title to calculate commodity association degree, so that the association degree calculation is very complex; in addition, the behavior characteristics that the user may buy the product after browsing and comparing are not considered, so that redundant comparison product data exists in the product with the shopping intention when the recently browsed product set is directly used as the shopping intention. The embodiment of the invention provides a commodity recommendation scheme mined based on a user behavior sequence to solve the technical problems in the prior art.
As shown in fig. 1, a commodity recommendation method mined based on a user behavior sequence according to an embodiment of the present invention mainly includes the following steps:
step S1: and collecting the behavior data of the user into a user behavior log in real time and formally describing the user behavior log.
When a user purchases online, the user can see a lot of goods to be selected, and when the goods are compared and decided, a series of behavior data can be generated for the goods, such as searching, browsing, collecting, adding a shopping cart, purchasing and the like. The series of behavior data is collected into a user behavior log in real time. Where real-time behavior data is communicated via a messaging system, such as the Kafka messaging system, and offline behavior data is stored in a database, such as the Hive database.
Table 1 exemplarily shows characteristic data of a user behavior log.
The data in table 1 contains 7 fields, respectively, session ID, action duration, user ID, action type, article ID, brand ID, category ID. The session ID is an ID of a single visit from the user logging in to the shopping website to logging out of the website, wherein the category ID refers to an ID of a category to which the commodity belongs, and the category is, for example, a jacket, trousers, hat or shoes, it should be noted that the category is not limited to a category in terms of wearing, that is, the category is also applicable to other categories. As the name implies, the brand ID is the ID of the brand to which the article belongs. And the article ID is a unique identification that distinguishes the article from other articles. For example, article a (sweater) and article B (shirt) are both jacket categories, i.e. they have the same category ID, but they have different article IDs to distinguish the sweater from the shirt.
The purpose of step S1 is to provide a data basis for subsequent real-time and offline data mining.
Step S2: and calculating the association degree of the commodity based on the behavior data in the offline shopping period.
Step S3: in the commodity set with the association degree larger than the predetermined threshold value, the intention of the user to purchase the commodity is calculated as the commodity to be recommended to the user based on the behavior data in the online shopping period.
The offline shopping period in the embodiment of the present invention may be, for example, one month. Incidentally, the offline shopping period is not limited to one month, but may be one or several weeks, or several months, or the like.
In the embodiment of the invention, the calculation of the association degree of the commodity is to express the shopping mode that the user watches the shopping, and the core of the calculation is to calculate the association degree of the commodity R (I)i，Ij). The association of the commodities comprises two meanings, namely commodity association of the same category, wherein the commodity relationship is used for recommending commodities with the same category ID but different commodity IDs to a user; and secondly, associating different categories of commodities, wherein the commodity relationship is used for recommending commodities which can be matched to the user. Thus, the degree of association of the merchandise may be defined as viewing the merchandise IiBut purchase commodity IjWhere i and j belong to natural numbers and i is not equal to j.
As shown in fig. 2, the steps of calculating the association degree of the product based on the offline shopping period by using the idea of frequent pattern mining are as follows:
step S21: and extracting browsing data of the user and purchasing data of the user from the behavior data in the offline shopping period according to the behavior type of the user. Wherein,
view ═ session ID, user ID, Commodity ID, Category ID
Buy ═ (Session ID, user ID, Commodity ID, Category ID)
Step S22: counting any commodity I in the browsing data of the useriThe number of occurrences N of each product ID is specifically counted. For example, it is implemented by HQL statement select commodity ID, count (session ID) from View group by commodity ID. The result is represented by (view. product ID, N), for example.
Step S23: and performing correlation calculation on the browsing data of the user and the purchasing data of the user to determine the similarity or matching degree of similar target commodities, wherein the join operation is performed on the browsing data of the user and the purchasing data of the user through an HQL statement. Wherein, the similarity of the same category commodities refers to that a user browses the commodity I from one visit from login to exit from a shopping websiteiBut purchase the same purpose goods IjThe commodity collocation degree refers to that the user browses the commodity I from one visit of logging in to exiting the shopping websiteiBut purchase goods I of a different purposejThe case (1). For example, the HQL language is used to describe, that is, when the join condition is: when (view, session ID ═ buy. session ID and view. category ID ═ buy. category ID) or (view, session ID ═ buy. session ID and view. category ID! ═ buy. category ID), the results after join are both expressed as (view. product ID, buy. product ID, 1).
Step S24: counting any one of the commodities I based on the result of the step S23iWith the purchased goods IjThe number of co-occurrences K is represented as (view. commodity ID, buy. commodity ID, K), for example.
Step S25: according to the formula R (I)i，Ij)＝P((Ij|Ii) Calculating the Commodity I as K/NiAnd IjI.e. browsing the item IiBut purchase the goods IjIs determined by the probability of
The result calculated in step S25 should be stored in a storage system of a recommendation object module, which will be explained in detail below.
As shown in fig. 3, the step of calculating the shopping intention product based on the online shopping cycle in step S3 of fig. 1 is as follows:
step S31: and obtaining a user behavior sequence in the online shopping period from the user behavior log. For example, aggregating the user behavior logs by session ID and user ID and forming a user behavior sequence, table 2 exemplarily shows how the user behavior sequence is formed,
session ID User ID Sequence of behaviors Final purchase
1 Uid1 S1,S2,S3,… I
2 Uid2 S1,S2,S3,… NULL
Therein, the behavior sequence of the client can be expressed as SiIs and SiIs any one of the sequence of actions and may be denoted SiTime of action, type of action, item ID, order of actionThe actions S1, S2 and S3 … in the column are arranged in ascending order according to their action durations, that is, the action duration of the first action S1 in the action sequence is the shortest, and the action durations of the following actions S2 and S3 … are sequentially increased. The last item purchased, I, may be a sequence of behaviors SiSome commodity is involved in but it could also be null.
Step S32: the preference function pre (I) for the last purchased commodity I is constructed in consideration of the frequency of occurrence of the last purchased commodity I in the user behavior sequence, the duration of the user's action on the last purchased commodity I, and the type of the user's behavior.
It can be clearly seen that:
a) the article ID of the last purchased article I is at { SiThe higher the frequency of appearance in the item I, the more preference the user has for the item I;
b) the action duration of the action performed by the user on the commodity I purchased last is set as SiThe larger the item I is, the more preference the user has for the item I;
c) the type of behavior of the user on the item I that was last purchased affects the user's preference for that item I.
Based on the above factors, assume { SiThe sequence length is N, the commodity I is in { S }iThe set of positions appearing (by way of the article ID) in (is { P } isi}，1≤PiN, the weight factor for a behavior type is expressed as WiIn which Σ Wi1. From the above information, for example, a calculation formula can be obtained:
wherein wpIndicates that position p corresponds toThe behavioral weight of.
Here, the set of positions { P }i}，1≤PiExemplary interpretation is made under N ≦ assuming a sequence of behaviors SiLength of 4, that is, a sequence of only four actions, i.e., S1, S2, S3, S4, and the article ID of the finally purchased article I is referred to in the first action S1 and the third action S3, then the set of positions { P of the finally purchased article I } is setiIs the sequence 1,3, i.e. the behavior belongs to the sequence of behaviors SiThe position in (c) }.
Step S33: and calculating the preference value of the last purchased commodity I in the user behavior sequence in the online shopping period, comparing the preference value with a preset preference threshold value, and selecting the commodity with the preference value larger than the preset preference threshold value as the shopping intention commodity of the user.
As shown in fig. 4, according to an embodiment of the present invention, there is provided a product recommendation device 40 mined based on a user behavior sequence, the product recommendation device including: the system comprises an obtaining behavior data module 401, a recommending object module 402 and a user interest module 403, wherein the obtaining behavior data module 401 is used for obtaining behavior data from the behavior of a user on a to-be-selected commodity and collecting the behavior data into a user behavior log in real time, the behavior data comprises behavior data in an offline shopping period and behavior data in an online shopping period, the recommending object module 402 calculates commodity association degrees based on the offline shopping period by adopting a frequent pattern mining idea, and the user interest module 403 calculates the commodity which the user intends to buy based on the online shopping period in a commodity set with the association degrees larger than a preset threshold value as a commodity to be recommended to the user.
The invention also provides an electronic device and a readable storage medium according to the embodiment of the invention.
The electronic device of the embodiment of the invention comprises: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executable by the at least one processor to cause the at least one processor to perform a method for recommending goods mined based on a sequence of user actions according to the present invention.
The computer-readable storage medium stores computer instructions for causing the computer to execute the commodity recommendation method based on user behavior sequence mining provided by the present invention.
Fig. 5 is a schematic hardware configuration diagram of an electronic device that executes a commodity recommendation method mined based on a user behavior sequence according to an embodiment of the present invention. As shown in fig. 5, the electronic apparatus includes: one or more processors 52 and a memory 51, one processor 52 being exemplified in fig. 5. The memory 51 is a computer-readable storage medium provided by the present invention.
The electronic device may further include: an input device 53 and an output device 54.
The processor 52, the memory 51, the input device 53 and the output device 54 may be connected by a bus or other means, as exemplified by the bus connection in fig. 5.
The memory 51, which is a computer-readable storage medium, may be used to store a software program, a computer-executable program, such as steps S1 to S3 of the commodity recommendation method mined based on the user behavior sequence in the embodiment of the present invention. The processor 52 executes the commodity recommendation method mined based on the user behavior sequence by executing software programs, instructions, and the like stored in the memory 51.
The input device 53 may receive the input access information. The output device 54 may include a display device such as a display screen.
The product can execute the method provided by the embodiment of the invention. For technical details that are not described in detail in this embodiment, reference may be made to the method provided by the embodiment of the present invention.
The above-described embodiments should not be construed as limiting the scope of the invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions can occur, depending on design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims (12)

1. A commodity recommendation method is characterized by comprising the following steps:
behavior data are obtained from behaviors of a user on a commodity to be selected and collected into a user behavior log in real time, wherein the behavior data comprise behavior data in an offline shopping period and behavior data in an online shopping period;
calculating the association degree of the commodity based on the behavior data in the offline shopping period;
and calculating the intention of the user to purchase the commodity as the commodity to be recommended to the user based on the behavior data in the online shopping period in the commodity set with the relevance degree larger than the preset threshold value.
2. The method of claim 1,
the behavioral data is data regarding at least one of searching, browsing, collecting, joining a shopping cart, and purchasing.
3. The method of claim 1, wherein the calculating the relevance of the good comprises:
according to the behavior type of the user, extracting browsing data of the user and purchasing data of the user from the behavior data in the offline shopping period;
counting any commodity I in the browsing data of the useriN;
performing correlation calculation on the browsing data of the user and the purchasing data of the user to determine the similarity of the commodities of the same category or the matching degree of the commodities, wherein the similarity of the commodities of the same category refers to the commodity I browsed by the user from logging in to exiting the shopping website in one visitiBut purchase the same purpose goods IjThe commodity collocation degree refers to that a user browses the commodity I from one visit of logging in to quitting the shopping websiteiBut purchase goods I of a different purposejThe case (1);
counting any commodity I according to the result of the correlation calculationiWith the purchased goods IjThe number of co-occurrences K; and
according to the formula R (I)i,Ij) Calculating the Commodity I as K/NiAnd IjAs the item I being browsediThereafter purchasing goods IjThe possibility of (a).
4. The method of claim 1, wherein the offline shopping period is one month.
5. The method of any of claims 1-4, wherein the calculating the user intent to purchase the item comprises:
obtaining a user behavior sequence in an online shopping period from the user behavior log;
the following preference function pre (I) for the last purchased item I is constructed,
suppose { SiThe length of the sequence is N, and the last purchased commodity I is in the sequence SiThe set of positions in is { P }i}，1≤PiN is less than or equal to N, the weight factor of the behavior type is { WiIn which Σ Wi＝1，
Wherein wpRepresenting the behavior weight corresponding to the position P;
comparing the calculated preference value with a preference threshold value and regarding the commodity of which the preference value is greater than the preference threshold value as a commodity which the user intends to shop.
6. A commodity recommending device based on user behavior sequence mining is characterized in that the commodity recommending device comprises: a behavior data obtaining module, a recommendation object module and a user interest module, wherein,
the behavior data obtaining module is configured to obtain behavior data from behaviors of the user on the to-be-selected commodity and collect the behavior data into a user behavior log in real time, wherein the behavior data comprises behavior data in an offline shopping period and behavior data in an online shopping period;
the recommendation object module is configured to calculate a commodity association degree based on behavior data of an offline shopping cycle, and
the user interest module is configured to calculate, based on an online shopping cycle, a user's intention to purchase a commodity as a commodity to be recommended to the user, among the set of commodities whose degree of association is greater than a predetermined threshold.
7. The item recommendation device of claim 6, wherein the behavior data is data regarding at least one of search, browse, favorite, join a shopping cart, and purchase.
8. The item recommendation device according to claim 6, wherein said calculating the association degree of the item comprises:
according to the behavior type of the user, extracting browsing data of the user and purchasing data of the user from the behavior data in the offline shopping period;
counting any commodity I in the browsing data of the useriN;
performing correlation calculation on the browsing data of the user and the purchasing data of the user to determine the similarity of the commodities of the same category or the matching degree of the commodities, wherein the similarity of the commodities of the same category refers to the commodity I browsed by the user from logging in to exiting the shopping website in one visitiBut purchase the same purpose goods IjThe commodity collocation degree refers to that a user browses the commodity I from one visit of logging in to quitting the shopping websiteiBut purchase goods I of a different purposejThe case (1);
counting any commodity I according to the result of the correlation calculationiWith the purchased goods IjThe number of co-occurrences K; and
according to the formula R (I)i，Ij) Calculating the Commodity I as K/NiAnd IjAs the item I being browsediThereafter purchasing goods IjThe possibility of (a).
9. The merchandise recommendation device of claim 6, wherein the offline shopping period is one month.
10. The item recommendation device according to any one of claims 6 to 9, wherein said calculating that the user intends to purchase an item comprises:
obtaining a user behavior sequence in an online shopping period from the user behavior log;
the following preference function pre (I) for the last purchased item I is constructed,
suppose { SiThe length of the sequence is N, and the last purchased commodity I is in the sequence SiThe set of positions in is { P }i}，1≤PiN is less than or equal to N, the weight factor of the behavior type is { WiIn which Σ Wi＝1，
Wherein wpRepresenting the behavior weight corresponding to the position p;
comparing the calculated preference value with a preference threshold value and regarding the commodity of which the preference value is greater than the preference threshold value as a commodity which the user intends to shop.
11. An electronic device, comprising:
at least one processor; and the number of the first and second groups,
a memory communicatively coupled to the at least one processor; wherein,
the memory stores instructions executable by the one processor to cause the at least one processor to perform the method of any one of claims 1-5.
12. A computer-readable storage medium storing computer instructions for causing a computer to perform the method of any one of claims 1-5.
CN201710206383.1A 2017-03-31 2017-03-31 Commodity recommendation method and device, electronic equipment and storage medium Active CN108665333B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710206383.1A CN108665333B (en) 2017-03-31 2017-03-31 Commodity recommendation method and device, electronic equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710206383.1A CN108665333B (en) 2017-03-31 2017-03-31 Commodity recommendation method and device, electronic equipment and storage medium

Publications (2)

Publication Number Publication Date
CN108665333A true CN108665333A (en) 2018-10-16
CN108665333B CN108665333B (en) 2021-04-30

Family

ID=63786854

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710206383.1A Active CN108665333B (en) 2017-03-31 2017-03-31 Commodity recommendation method and device, electronic equipment and storage medium

Country Status (1)

Country Link
CN (1) CN108665333B (en)

Cited By (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109961327A (en) * 2019-04-11 2019-07-02 上海拉扎斯信息科技有限公司 Data processing method, device, electronic equipment and computer readable storage medium
CN110060099A (en) * 2019-04-04 2019-07-26 北京字节跳动网络技术有限公司 Information-pushing method, device and electronic equipment based on historical record
CN110689401A (en) * 2019-08-30 2020-01-14 江苏苏宁物流有限公司 Service commodity recommendation method and device
CN110996110A (en) * 2019-12-02 2020-04-10 深圳市云积分科技有限公司 Commodity adjusting method and device in live broadcast process
CN111161021A (en) * 2019-12-23 2020-05-15 叮当快药科技集团有限公司 Real-time feature-based quick secondary sorting method and tool for recommended commodities
CN111461841A (en) * 2020-04-07 2020-07-28 腾讯云计算（北京）有限责任公司 Article recommendation method, device, server and storage medium
CN111563791A (en) * 2020-03-31 2020-08-21 北京奇艺世纪科技有限公司 Commodity dividing and recommending method and device, electronic equipment and storage medium
CN111612557A (en) * 2019-02-25 2020-09-01 阿里巴巴集团控股有限公司 Method and device for providing commodity object information and electronic equipment
CN111695917A (en) * 2019-03-11 2020-09-22 北京京东尚科信息技术有限公司 Commodity recommendation method, system, electronic device and storage medium
CN111985994A (en) * 2020-08-06 2020-11-24 上海博泰悦臻电子设备制造有限公司 Commodity recommendation method and related equipment
CN112862567A (en) * 2021-02-25 2021-05-28 华侨大学 Exhibit recommendation method and system for online exhibition
CN112862540A (en) * 2021-03-08 2021-05-28 重庆第二师范学院 Advertisement putting method and device based on big data, storage medium and server
CN113065932A (en) * 2021-05-06 2021-07-02 北京京东振世信息技术有限公司 Article recommendation method and device
CN113761002A (en) * 2021-03-25 2021-12-07 北京沃东天骏信息技术有限公司 Information pushing method, device, equipment and computer readable storage medium
WO2022116833A1 (en) * 2020-12-01 2022-06-09 北京沃东天骏信息技术有限公司 Product search method, terminal device, and server
CN116911960A (en) * 2023-09-12 2023-10-20 贵昌集团有限公司 Electronic commerce data recommendation method based on big data
CN117035948A (en) * 2023-10-10 2023-11-10 山东唐和智能科技有限公司 Task intelligent processing method and system based on big data
CN117557318A (en) * 2023-12-29 2024-02-13 青岛巨商汇网络科技有限公司 Management intelligent analysis method and system based on virtual shopping

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101206752A (en) * 2007-12-25 2008-06-25 北京科文书业信息技术有限公司 Electric commerce website related products recommendation system and method
CN102346894A (en) * 2010-08-03 2012-02-08 阿里巴巴集团控股有限公司 Output method, system and server of recommendation information
CN102479366A (en) * 2010-11-25 2012-05-30 阿里巴巴集团控股有限公司 Commodity recommending method and system
CN105528374A (en) * 2014-10-21 2016-04-27 苏宁云商集团股份有限公司 A commodity recommendation method in electronic commerce and a system using the same
CN105589905A (en) * 2014-12-26 2016-05-18 中国银联股份有限公司 User interest data analysis and collection system and method

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101206752A (en) * 2007-12-25 2008-06-25 北京科文书业信息技术有限公司 Electric commerce website related products recommendation system and method
CN102346894A (en) * 2010-08-03 2012-02-08 阿里巴巴集团控股有限公司 Output method, system and server of recommendation information
CN102479366A (en) * 2010-11-25 2012-05-30 阿里巴巴集团控股有限公司 Commodity recommending method and system
CN105528374A (en) * 2014-10-21 2016-04-27 苏宁云商集团股份有限公司 A commodity recommendation method in electronic commerce and a system using the same
CN105589905A (en) * 2014-12-26 2016-05-18 中国银联股份有限公司 User interest data analysis and collection system and method

Cited By (28)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111612557B (en) * 2019-02-25 2023-06-30 阿里巴巴集团控股有限公司 Method and device for providing commodity object information and electronic equipment
CN111612557A (en) * 2019-02-25 2020-09-01 阿里巴巴集团控股有限公司 Method and device for providing commodity object information and electronic equipment
CN111695917A (en) * 2019-03-11 2020-09-22 北京京东尚科信息技术有限公司 Commodity recommendation method, system, electronic device and storage medium
CN110060099B (en) * 2019-04-04 2021-07-20 北京字节跳动网络技术有限公司 Information pushing method and device based on historical records and electronic equipment
CN110060099A (en) * 2019-04-04 2019-07-26 北京字节跳动网络技术有限公司 Information-pushing method, device and electronic equipment based on historical record
CN109961327A (en) * 2019-04-11 2019-07-02 上海拉扎斯信息科技有限公司 Data processing method, device, electronic equipment and computer readable storage medium
CN110689401A (en) * 2019-08-30 2020-01-14 江苏苏宁物流有限公司 Service commodity recommendation method and device
CN110996110A (en) * 2019-12-02 2020-04-10 深圳市云积分科技有限公司 Commodity adjusting method and device in live broadcast process
CN110996110B (en) * 2019-12-02 2022-02-18 深圳市云积分科技有限公司 Commodity adjusting method and device in live broadcast process
CN111161021B (en) * 2019-12-23 2021-10-08 叮当快药科技集团有限公司 Quick secondary sorting method for recommended commodities based on real-time characteristics
CN111161021A (en) * 2019-12-23 2020-05-15 叮当快药科技集团有限公司 Real-time feature-based quick secondary sorting method and tool for recommended commodities
CN111563791B (en) * 2020-03-31 2023-12-26 北京奇艺世纪科技有限公司 Commodity dividing and recommending method and device, electronic equipment and storage medium
CN111563791A (en) * 2020-03-31 2020-08-21 北京奇艺世纪科技有限公司 Commodity dividing and recommending method and device, electronic equipment and storage medium
CN111461841B (en) * 2020-04-07 2023-04-07 腾讯云计算（北京）有限责任公司 Article recommendation method, device, server and storage medium
CN111461841A (en) * 2020-04-07 2020-07-28 腾讯云计算（北京）有限责任公司 Article recommendation method, device, server and storage medium
CN111985994A (en) * 2020-08-06 2020-11-24 上海博泰悦臻电子设备制造有限公司 Commodity recommendation method and related equipment
WO2022116833A1 (en) * 2020-12-01 2022-06-09 北京沃东天骏信息技术有限公司 Product search method, terminal device, and server
CN112862567A (en) * 2021-02-25 2021-05-28 华侨大学 Exhibit recommendation method and system for online exhibition
CN112862540B (en) * 2021-03-08 2022-09-13 重庆第二师范学院 Advertisement putting method and device based on big data, storage medium and server
CN112862540A (en) * 2021-03-08 2021-05-28 重庆第二师范学院 Advertisement putting method and device based on big data, storage medium and server
CN113761002A (en) * 2021-03-25 2021-12-07 北京沃东天骏信息技术有限公司 Information pushing method, device, equipment and computer readable storage medium
CN113065932A (en) * 2021-05-06 2021-07-02 北京京东振世信息技术有限公司 Article recommendation method and device
CN116911960A (en) * 2023-09-12 2023-10-20 贵昌集团有限公司 Electronic commerce data recommendation method based on big data
CN116911960B (en) * 2023-09-12 2023-11-17 贵昌集团有限公司 Electronic commerce data recommendation method based on big data
CN117035948A (en) * 2023-10-10 2023-11-10 山东唐和智能科技有限公司 Task intelligent processing method and system based on big data
CN117035948B (en) * 2023-10-10 2024-01-09 山东唐和智能科技有限公司 Task intelligent processing method and system based on big data
CN117557318A (en) * 2023-12-29 2024-02-13 青岛巨商汇网络科技有限公司 Management intelligent analysis method and system based on virtual shopping
CN117557318B (en) * 2023-12-29 2024-06-11 青岛巨商汇网络科技有限公司 Management intelligent analysis method and system based on virtual shopping

Also Published As

Publication number Publication date
CN108665333B (en) 2021-04-30

Similar Documents

Publication Publication Date Title
CN108665333B (en) Commodity recommendation method and device, electronic equipment and storage medium
CN110162693B (en) Information recommendation method and server
CN108665329A (en) A kind of Method of Commodity Recommendation based on user browsing behavior
US10664888B2 (en) Method and system for attribute extraction from product titles using sequence labeling algorithms
CN110020128B (en) Search result ordering method and device
CN103400286B (en) A kind of commending system and method carrying out article characteristics mark based on user behavior
US9400831B2 (en) Providing information recommendations based on determined user groups
US9727620B2 (en) System and method for item and item set matching
CN112132660B (en) Commodity recommendation method, system, equipment and storage medium
CN109087177A (en) To the method, apparatus and computer readable storage medium of target user's Recommendations
US11748799B2 (en) Method, medium, and system for information recommendation
CN104317945A (en) E-commerce website commodity recommending method on basis of search behaviors
CN102609422A (en) Class misplacing identification method and device
CN112036981B (en) Method, device, equipment and medium for providing target comparison commodities
Lestari et al. Performance comparison of rank aggregation using borda and copeland in recommender system
CN108733705A (en) A kind of effective sequential mode mining method and device
CN113689258A (en) Commodity recommendation heat generation method and system
Chen et al. Product recommendation system for small online retailers using association rules mining
Monteserin et al. Influence-based approach to market basket analysis
CN113225580B (en) Live broadcast data processing method and device, electronic equipment and medium
CN110765346B (en) User intention mining method, device and equipment
KR101637986B1 (en) System for providing the optimal products and method thereof
CN112184250A (en) Method and device for generating retrieval page, storage medium and computer equipment
Verma et al. Collaborative filtering with label consistent restricted boltzmann machine
CN114240545A (en) Recommendation method, device and equipment for similar commodities and storage medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant